Unicode Detector and
Hidden Symbol Finder.

Paste any text to instantly find and remove invisible Unicode characters: zero-width spaces, directional marks, smart quotes, and hidden formatting that breaks APIs, publishing tools, and systems.

  • No sign-up required
  • No limits
  • Runs in your browser

Detection

What We Detect.

The scanner identifies four categories of problematic characters and normalizes each to its ASCII equivalent or removes it entirely.

Invisible

Invisible Characters

Characters with no visual representation: zero-width spaces, directional marks, word joiners. Invisible in plain text but present in the data, causing issues with search, comparison, and pasting.

  • U+200B Zero-width space
  • U+200D Zero-width joiner
  • U+00AD Soft hyphen
Spaces

Non-standard Spaces

Characters that look like regular spaces but behave differently, causing alignment issues, broken word wrapping, and text comparison failures.

  • U+00A0 Non-breaking space
  • U+2003 Em space
  • U+3000 Ideographic space
Punctuation

Smart Punctuation

Typographic characters that look elegant in print but break code, APIs, and systems expecting standard ASCII: curly quotes, em dashes, and the ellipsis character.

  • U+201C/201D Curly quotes
  • U+2014 Em dash
  • U+2026 Ellipsis
Control

Control Characters

Characters that control text direction, rendering, or Unicode variant selection. Often copied unnoticed from rich text. Includes BiDi overrides, variation selectors, and format codes.

  • U+202E Right-to-left override
  • U+FE0F Variation selector
  • U+2066 to 2069 Isolates

Why It Matters

Text That Looks Clean
Often Isn't.

Content copied from AI tools, PDFs, or rich editors routinely carries invisible characters. They are harmless to the naked eye but cause real problems: broken JSON, inconsistent search results, encoding errors in databases, and failed comparisons.

APIs and JSON

Hidden characters can break parsers, invalidate strings, and cause silent data corruption.

Databases

Non-standard characters lead to encoding mismatches and inconsistent query results.

Publishing

CMS platforms, email tools, and similar systems may render smart punctuation incorrectly.

AI Detection

Some detectors treat unusual Unicode as a weak signal. Clean text produces consistent, predictable results.

Origins

Where Invisible Characters Come From.

Invisible characters rarely appear on their own. They arrive when text moves between tools, and each source leaves its own fingerprint. Knowing where they come from makes it easier to know when to check for hidden characters before content reaches a system that cannot handle them.

Rich Text Editors.

Word processors and design tools store formatting alongside the text itself. Copy a paragraph from Word, Google Docs, or Figma and non-breaking spaces, soft hyphens, and smart punctuation travel with it. The pasted text looks identical to the original, which is why few people think to check text for invisible characters until a form rejects the input or a search fails to match. A quick scan with a Unicode detector shows everything the editor added.

AI Tools and Chat Interfaces.

Chat interfaces render responses as rich text. Copying output from ChatGPT, DeepSeek, or any assistant can carry zero-width characters, variation selectors, and typographic punctuation into your document. The characters survive pasting into most editors and travel into whatever you publish next. Running AI output through a character scanner before it goes live catches these leftovers in seconds.

PDFs and Exports.

PDF is a layout format, not a text format. Extracting text from one introduces ligatures, hyphenation artifacts, and unusual space characters that were never in the source document. Exports from email clients and CMS platforms behave the same way. If content passed through a PDF at any point, treat it as unclean until a character detector says otherwise.

The fix is a habit rather than a workflow change. Check for invisible characters whenever text crosses a system boundary: before it enters a database, an API request, a CMS, or a published page. The scanner above runs the full check in your browser, flags every hidden character it finds, and cleans them in one click. No text leaves your device.

FAQ

Questions &
Answers.

Everything you need to know before you start.

4.8 / 5

Content That Sounds Human.
Published Faster.

Start with 1,000 free words. No credit card. See the result in seconds.

Before AI Draft
84% AI likelihood
After Human
11% AI likelihood
Try HumanTone Free
1,000 free words No credit card Cancel anytime